The blood-brain barrier (BBB), a specialization of the neurovasculature that normally protects CNS tissues against immune cell invasion, and the limited ability to replace damaged neurons represent two major obstacles in clearing a neurotropic virus from the CNS. Although the mechanisms responsible for immune clearance of neurotropic virus are poorly described, viral encephalitis often resolves without permanent disability. This is not the case for infection with wild-type rabies virus (RV) where the outcome is almost always lethal if the virus reaches the CNS. Even where the RV vaccine and pre-formed virus-neutralizing antibodies (VNA) used in post- exposure prophylaxis (PEP) are available, human rabies has re-emerged. This is primarily due to infection with RV carried by bats, which is often not recognized until the virus has reached the CNS and cannot be cleared by the current PEP regimen. The recognition that many laboratory RV strains developed as live-attenuated vaccines spread to, and replicate in the CNS tissues but are cleared by immune mechanisms without overt neuropathology has led to renewed interest in why natural RV infections are not cleared from the CNS. Comparative studies of infection of mice with different RV have revealed that the defect in the immune response to wildlife strains is in the interaction between invading immune cells and cells of the neurovascular unit (NVU), including vascular endothelial cells, astrocytes and other cells that make up the BBB. In lethal rabies BBB integrity is maintained and immune effectors capable of clearing the virus cannot enter infected CNS tissues. In this project the immune-NVU interactions responsible for the delivery of immune effectors across the BBB in response to RV infection and the immune effectors that clear RV will be characterized using a variety of molecular and immunological approaches including established adoptive transfer protocols. Then the utility of novel, reverse-engineered recombinant RV vaccines in overcoming the block on clearance of lethal wildlife RV from the CNS will be assessed. These experiments will be performed in mouse models since RV infect all mammals and mice, like humans, are end-stage hosts of the virus that exhibit many of the features of human rabies. The aim is to understand how a neurotropic virus can be cleared from the CNS with minimal pathology.